Mucoepidermoid carcinoma: a clinicopathologic study of 80 patients with special reference to histological grading.

We sought to review our experience with salivary mucoepidermoid carcinoma (MEC) over two decades to confirm the validity and reproducibility of histologic grading and to investigate MIB-1 index as a prognosticator. Diagnosis was confirmed on 80 cases, and chart review or patient contact was achieved for 48 patients, with follow-up from 5 to 240 months (median 36 months). Immunohistochemistry with citrate antigen retrieval for MIB-1 was performed on a subset of cases. Kaplan-Meier survival curves were generated for each stage, site, and grade according to our proposed grading system. To address the issue of grading reproducibility, 20 slides were circulated among five observers, without prior discussion; slides were categorized as low-, intermediate-, or high-grade according to one's "own" criteria, and then according to the AFIP criteria proposed by Goode et al.10 Weighted kappa (kappa) estimates were obtained to describe the extent of agreement between pairs of rating. The Wilcoxon signed rank test or the Friedman test as appropriate tested variation across ratings. There was no gender predominance and a wide age range (15-86 years, median 49 years). The two most common sites were parotid and palate. All grade 1 MECs presented as Stage I tumors, and no failures were seen for this category. The local disease failure rates at 75 months for grades 2 and 3 MEC were 30% and 70%, respectively. Tumor grade, stage, and negative margin status all correlated with disease-free survival (DFS) (p = 0.0091, 0.0002, and 0.048, respectively). The MIB index was not found to be predictive of grade. Regarding the reproducibility of grading, the interobserver variation for pathologists using their "own" grading, as expressed by the kappa value, ranged from good agreement (kappa = 0.79) to poor (kappa = 0.27) (average kappa = 0.49). A somewhat better interobserver reproducibility was achieved when the pathologists utilized the standardized AFIP criteria (average kappa = 0.61, range 0.38-0.77). This greater agreement was also reflected in the Friedman test (statistical testing of intraobserver equality), which indicated significant differences in using one's own grading systems (p = 0.0001) but not in applying the AFIP "standardized" grading (p = 0.33). When one's own grading was compared with the AFIP grading, there were 100 pairs of grading "events," with 46 disagreements/100 pairs. For 98% of disagreements, the AFIP grading "downgraded" tumors. This led us to reanalyze a subset of 31 patients for DFS versus grade, for our grading schema compared with the AFIP grading. Although statistical significance was not achieved for this subset, the log rank value revealed a trend for our grading (p = 0.0993) compared with the Goode schema (p = 0.2493). This clinicopathologic analysis confirms the predictive value of tumor staging and three-tiered histologic grading. Our grading exercise confirms that there is significant grading disparity for MEC, even among experienced ENT/oral pathologists. The improved reproducibility obtained when the weighted AFIP criteria were used speaks to the need for an accepted and easily reproducible system. However, these proposed criteria have a tendency to downgrade MEC. Therefore, the addition of other criteria (such as vascular invasion, pattern of tumor infiltration [i.e., small islands and individual cells vs cohesive islands]) is necessary. We propose a modified grading schema, which enhances predictability and provides much needed reproducibility.

Sinonasal undifferentiated carcinoma: immunohistochemical profile and lack of EBV association.

The role of Epstein-Barr virus (EBV) in the development of sinonasal undifferentiated carcinoma (SNUC) remains unresolved. Reports of EBV-positivity in SNUC may reflect inclusion of lymphoepithelioma-like carcinomas within this group. In addition, SNUC have been incompletely characterized immunohistochemically, and their undifferentiated appearance often requires such ancillary studies to aid in their distinction from other high-grade neoplasms. To address these two issues, 25 cases of SNUC diagnosed between the years 1983 and 1999 were selected from our files. EBER in situ hybridization (ISH) was performed on the paraffin-embedded tissue by using 3H-labeled EBER-1 RNA probes. Neoplasms with sufficient tissue (22 of 25) were evaluated immunohistochemically for Ki-67, p53, chromogranin, synaptophysin, placental alkaline phosphatase (PLAP), CD99, carcinoembryonic antigen (CEA), epithelial membrane antigen (EMA), neuron-specific enolase (NSE), and latent membrane protein-1 (LMP-1). The median patient age was 58 years (range, 20-81 years), with a male-to-female ratio of approximately 3:1. The most common tumor location was the nasal cavity (18 cases), followed by the ethmoid and maxillary sinuses. Median survival was 18 months. All 25 tumors were negative for EBER-I by ISH. Ki-67 was negative in one case, 1+ in nine, 2+ in six, 3+ in five, and 4+ in one. P53 was negative in nine, 1+ in five, 2+ in two, 3+ in none, and 4+ in six. CD99 expression was strongly positive in 3 of 22 (14%) and completely negative in the remainder. Variably intense focal staining for EMA was present in 4 of 22 (18%). NSE faintly stained 4 of 22 (18%). Chromogranin, synaptophysin, PLAP, CEA, and LMP-1 were negative (0 of 22). Our results suggest that EBV does not play a role in the development of SNUC. Strict histologic criteria are necessary to avoid confusion with lymphoepithelioma-like carcinoma or other high-grade malignancies in this region. The finding of occasional CD99-positive cases adds SNUC to the growing list of CD99-positive neoplasms.

Use of the peritracheal fold in the dog tracheal transplantation model.

OBJECTIVE: To investigate the technical aspects of the canine model of human tracheal transplantation for potential application to reconstruction of extremely long tracheal defects (> 10 cm). DESIGN: In phase 1, long tracheal segments were skeletonized and pedicled with the thyroid glands, cranial thyroid arteries and veins, and internal jugular vein branches. The segments were elevated completely, attached to the vascular pedicle only, and replaced with primary tracheal anastomoses. In phase 2, long segments were elevated along with a diffuse soft tissue "blanket" that envelops the trachea and thyroid glands. Because this study was designed to primarily address, in situ, tracheal perfusion territories of a cranially located vascular pedicle, microvascular anastomoses were not conducted. SUBJECTS: Two small-bodied beagles (10-15 kg) and 5 large-bodied mixed-breed dogs (20-30 kg) were humanely killed 2 to 41 days after surgery, and anatomic and histological analyses were conducted. RESULTS: Unlike that of humans, the thyroid gland complex of dogs is not intimately associated with the trachea but is conjoined with a peritracheal soft tissue "fold." Within this fold, blood is transmitted to the trachea via a diffuse, segmental vascular plexus. In phase 1, pronounced tracheal necrosis occurred within 2 to 5 days. In phase 2, extremely long tracheal segments (10-12 cm), based only on a cranially located pedicle, were still viable at 2 to 6 weeks. CONCLUSIONS: Preservation of the "peritracheal fold" in the dog model of tracheal transplantation is critical to the onset and maintenance of vascular perfusion in a long tracheal segment. Furthermore, the use of large-bodied dogs is necessary to provide for a usable venous efflux component.

Upper aerodigestive tract liposarcoma: report on four cases and literature review.

OBJECTIVE: To report on the clinical behavior, histopathology, treatment, and prognosis of laryngeal, hypopharyngeal, and cervical esophageal liposarcomas. STUDY DESIGN: Retrospective reviews of pathology files and hospital records at a tertiary care hospital and a retrospective search of the English-language literature. METHODS: Cases of upper aerodigestive tract (UADT) liposarcoma with adequate histopathologic documentation and clinical information were included for review. RESULTS: Four cases of UADT liposarcomas were identified. The literature review revealed 26 cases of laryngeal liposarcomas, 7 cases of hypopharyngeal liposarcomas, and 6 cases of esophageal liposarcomas: the mean age at presentation was 55.8 years, the male:female ratio was 5:1, and 60% of the patients presented with dysphagia. Eighty-six percent of tumors had low-grade histologic findings. The recurrence rate after primary resection was 50%. Recurrence correlated with surgical procedure rather than with histologic subtype; 94.7% of recurrences happened after simple excision. Distant metastases occurred in three patients; two of them died of the disease. CONCLUSIONS: The literature supports that UADT liposarcomas are rare and usually of low-grade histologic type. The rate of metastatic disease and tumor-related mortality is low. However, high recurrence rates have been noted, particularly when less radical surgery is employed.

Human papillomavirus may be common within nasopharyngeal carcinoma of Caucasian Americans: investigation of Epstein-Barr virus and human papillomavirus in eastern and western nasopharyngeal carcinoma using ligation-dependent polymerase chain reaction.

BACKGROUND: Nasopharyngeal carcinoma (NPC), particularly those tumors endemic to the Far East, commonly harbor Epstein-Barr virus (EBV), thought to serve as an important oncogenic promoter. Human papillomavirus (HPV) is associated with a proportion of upper aerodigestive tract carcinomas. We hypothesized that HPV might also contribute to the pathogenesis of NPC, and we queried whether geographic and racial distinctions may be identified between NPC of the Far East versus those diagnosed in Caucasian American patients with regard to the interrelationship of histologic subtype and viral infection. MATERIALS AND METHODS: Formalin-fixed paraffin-embedded tissue (FFPET) from 30 patients (6 Caucasian Americans, 1 Chinese American, 14 and 9 patients from Korea and China, respectively) were studied using the ligation-dependent polymerase chain reaction (LD-PCR). These cases were histologically classified according to the World Health Organization (WHO) schema for NPC. Consensus target probes complementary to the L1 region of over 30 HPV types, as well as target probes complementary to EBER-1 (EBV-related nontranslated latency-associated RNA), were used to amplify target sequences. RESULTS: Seven of 30 NPC (23%) contained HPV sequences. There were 6 Caucasian American patients with NPC; 3 cases (50%) were HPV positive (HPV+). Two of these Caucasian Americans had WHO type I tumors: one was HPV+ and EBV negative (EBV-) and the other was HPV-/EBV+. The remaining Caucasian American NPCs were WHO-II/III tumors which tested as follows: two were coinfected with HPV and EBV, the other two contained EBER but not HPV sequences. The single Oriental American patient had a WHO-III NPC which was HPV-/EBV+. Of the Eastern NPC patients, 4 (1 WHO-I, 3 WHO-II/III) of 23 (17%) NPCs contained HPV sequences as well as EBV. Conclusion. Human papillomavirus appears to be uncommonly (17%) associated with NPC in patients from the Far East and was detected more often (50%) in NPC from American Caucasian patients. Some of these tumors conformed to our perceptions and expectations of NPC (eg, WHO-I tumors being EBV-/HPV+ and WHO-III tumors being EBV+/HPV-), but other tumors did not conform to these expectations (eg, WHO-III NPC occasionally harboring both HPV and EBV). There appears to be a broad profile in the relationship between HPV, EBV, and NPC histologic subtype. Unfortunately, the number of American Caucasian cases studied are too small to allow for strong conclusions.

Characterizing the antigenic profile of the human trachea: implications for tracheal transplantation.

BACKGROUND: Tracheal transplantation may be a viable alternative in select situations of long-segment tracheal stenosis. Issues concerning human tracheal antigenicity and the requirement for systemic immunosuppression need to be addressed. This study examined the distribution of the major transplantation antigens on fresh human trachea. METHODS: Frozen sections of 10 human tracheas were examined for class I (HLA [human leukocyte antigen]-A, -B, -C) and class II (HLA-DR) histocompatibility antigens using mouse monoclonal antibodies in an indirect immunoperoxidase assay. The tracheal mucosa, perichondrium, cartilage matrix, and chondrocytes were examined and assigned a semiquantitative score for HLA class I and class II expression. RESULTS: The tracheal mucosa was strongly antigenic for HLA class I and class II antigens, whereas the perichondrium contained only scattered fibroblasts that were positive for both classes. The polysaccharide matrix was consistently devoid of class I or class II expression. The chondrocytes of all 10 specimens expressed diffuse class I positivity; only two specimens contained chondrocytes that expressed class II positivity. CONCLUSIONS: The human trachea contains "compartments" of highly antigenic tissue that are morphologically distinct from regions of lesser antigenicity.

Sinonasal melanoma. A clinicopathologic study of 25 cases and literature meta-analysis.

OBJECTIVES: To report on 25 cases of sinonasal melanoma at our institution during 1977 to 1995 and to perform a literature meta-analysis on sinonasal melanoma and study survival statistics. DATA SOURCES: A retrospective search of pathology and clinical records at our institution to identify patients with sinonasal melanoma; a retrospective search of the English-language literature for cases of documented sinonasal melanoma. STUDY SELECTION: Cases from our institution were included if there was adequate pathologic material confirming the pathologic diagnosis. For the meta-analysis, cases were accepted if they included convincing histologic illustrations or documentation of melanin by Fontana-Masson staining or ultrastructural examination or had an acceptable immunohistochemical profile with some documented clinical follow-up on individual patients. DATA EXTRACTION: The following parameters were monitored for patients from our institution: age, sex, possible exposure histories, tumor site, tumor size, type of therapy, clinical outcome, cytochemistry (Fontana-Masson stain), and immunohistochemical profile (S-100, HMB-45, vimentin, cytokeratin). The parameters of the meta-analysis were age, sex, tumor site, and survival data. DATA SYNTHESIS: Our patient population included 10 men and 15 women aged 23 to 83 years (mean, 65 years). Tumor sites included inferior turbinate, superior nasal cavity, nasal cavity floor/palate, ethmoid sinuses, and maxillary sinus. The immunohistochemical profile was as follows (positive cases): S-100 (14 of .14), HMB-45 (12 of 14), vimentin (9 of 10), cytokeratin (0 of 9). Melanin was demonstrated by Fontana-Masson stain (8 of 8 cases). Follow-up information was available in 25 cases, with follow-up ranging from 1 to 217 months (mean, 39 months). Eleven patients were disease free at 1 to 217 months (mean, 59 months) after excision. Local single recurrences developed in 6 patients at 3 to 84 months (mean, 14 months), and were treated with surgery. Multiple local recurrences were seen in 2 patients. One of them died of disseminated disease, after 2 local recurrences, 60 months after original diagnosis. The other was disease free after 38 months. Ten patients ultimately died of disease within 2 to 60 months after diagnosis (mean, 21 months). Only 2 patients may have developed metastatic disease in the absence of local recurrence. A total of 163 cases were identified from the literature: 53 women, 54 men (56 gender unknown). Stratification by era (pre-1980 vs post-1980) did not reveal any gender differences (pre-1980: 34 women [48%], 37 men [52%]; post-1980: 34 women [55%], 27 men [44%]). However, stratification by era did reveal an age shift toward older patients in the 1980-1995 era. The 5-year median survival for all patients was 36 months. Stratification by era of treatment revealed a long-term survival trend: patients treated before 1980 had somewhat better survival at 60 and 100 months (40% and 36%, respectively) than patients diagnosed and treated between 1980 and 1995 (30% and 23%, respectively), although statistical significance was not reached. CONCLUSION: It is unclear why the long-term survival for patients with sinonasal melanoma seems to be decreasing. Despite advances in imaging techniques, which would potentially allow for earlier tumor discovery, and advances in surgical techniques and adjuvant therapeutics, the mean survival for patients with sinonasal melanoma has not improved during the past 15 years.

Osteogenic sarcoma of the jaw. A clinicopathologic review of 12 patients.

OBJECTIVE: To review our experience with osteogenic sarcoma of the jaw. DATA SOURCE: The pathology and clinical files of the Mount Sinai Medical Center, New York, NY, were examined. STUDY SELECTION: For the past 6 years, 12 pathologically documented cases were identified. DATA EXTRACTION: Surgical resection margins, histological tumor grade and subtype, and treatment modality were reexamined with regard to effect on outcome. DATA SYNTHESIS: The mean age of occurrence was 32.6 years, ranging from 12 to 60 years. Male predominance was noted in both mandibular and maxillary tumors. Local recurrence occurred in 6 of 12 patients after 3 months to 5 years; 4 of these patients have either persistent disease or died of disease. In this recurrence group, only 2 patients remain disease free 7 and 12 years after initial diagnosis and 6.5 and 8 years after recurrence. CONCLUSIONS: This study is too small to draw statistically significant conclusions. However, the longest survivor in the group (disease free at 12 years) had a histologically low-grade mandibular tumor. We reiterate the findings in the literature, which agree that osteogenic sarcoma of the jaw is an aggressive disease. Patient prognosis may depend on the tumor grade and size on presentation, as well as the completeness of initial surgery as monitored by resection margins.

Nodal inclusion cysts of the parotid gland and parapharyngeal space: a discussion of lymphoepithelial, AIDS-related parotid, and branchial cysts, cystic Warthin's tumors, and cysts in Sjögren's syndrome.

The purpose of this report is to examine the computed tomography scans, magnetic resonance images, and pathologic findings in 44 patients, 42 of whom had inclusion-type cysts of the parotid and parapharyngeal space of varying etiologies. Two additional cases of cystic changes in the benign lymphoepithelial lesion (BLEL) of Sjögren's syndrome are highlighted here, since they had unusually large cystic components mimicking acquired immunodeficiency syndrome-related parotid cysts (ARPCs). A retrospective examination identified 18 ARPCs, 3 lymphoepithelial cysts (LECs), 13 cystic Warthin's tumors, 8 branchial cysts, and 2 cases of cysts in patients with Sjögren's syndrome (BLEL), all of whom had imaging studies and pathologic confirmation. There were 30 men and 14 women with an age range of 25 to 72 years (median, 46.82 years). Any similarities in the imaging appearances were noted, as were any differences in pathologic detail. On imaging, only the cystic Warthin's tumors had any focal wall nodularity; the other cysts had smooth walls. When multiple parotid cysts were present, the distinguishing feature between ARPCs and cysts in BLEL (and some cystic Warthin's tumors) was the presence of diffuse cervical adenopathy in patients with ARPCs. Imaging usually could not differentiate between a solitary parotid LEC, a branchial cyst, and some cystic Warthin's tumors. Extraparotid lesions were either branchial cysts or cystic Warthin's tumors. Physicians should be aware of the variety of different inclusion-type cysts that may occur in the parotid gland and parapharyngeal space, all of which may have similar imaging appearances. Although imaging clearly identifies these cysts and may suggest a specific diagnosis, it must always be remembered that the precise diagnosis remains in the province of the pathologist.

Inflammatory pseudotumor of the maxillary sinus: CT and MR findings in six cases.

OBJECTIVE: Inflammatory pseudotumor is a chronic inflammatory lesion of unknown origin. The purpose of this study was to analyze the CT and MR findings in six cases of inflammatory pseudotumor of the maxillary sinus in order to determine if there were any imaging findings that distinguish this lesion from the more common type of pseudotumor that occurs in the orbit. MATERIALS AND METHODS: Six patients with pseudotumor of the maxillary sinus were identified retrospectively. The diagnosis was based on biopsy specimens. Their CT scans and MR images were studied by two radiologists to identify specifically the presence of soft-tissue disease and the presence and type of any bone involvement. RESULTS: CT and MR images showed some degree of involvement of the bony wall in all cases. All patients had a soft-tissue mass with mild enhancement on CT scans and intermediate signal intensity on both T1- and T2-weighted MR images. CONCLUSION: Our experience shows that inflammatory pseudotumor of the maxillary sinus causes bone changes on CT and MR images, findings that mimic a malignant tumor. Bone involvement appears to be more common in pseudotumor of the maxillary antrum than in its orbital counterpart.

Analysis of prevalence of human papillomavirus in laryngeal carcinomas. Study of 40 cases using polymerase chain reaction and consensus primers.

In this study we determined the frequency of association of human papillomavirus (HPV) and laryngeal carcinomas and investigated the possibility that HPV may be associated with larger or more aggressive tumors. Laryngeal squamous cell carcinomas from 40 patients who did not have preexisting papillomas by clinical history were retrieved from formalin-fixed, paraffin-embedded blocks and analyzed for HPV. Twenty-two cases were tumors of the true vocal folds, and 18 were supraglottic. Clinical follow-up was available for 25 patients. We used the polymerase chain reaction (PCR) with the "hot start" modification and consensus primers that can detect over 30 distinct HPV types. Three of the 40 patients (8%) had detectable HPV DNA. These 3 patients did not have unusual age demographics and were smokers. All but 1 of the 22 HPV-negative patients who were questioned were also smokers. We compared the outcomes for large (4 cm or greater) HPV-positive and -negative tumors. Six of the 40 tumors were 4 cm or greater and involved contiguous structures. Two of these 6 were HPV-positive, and these patients died of disease after 3 and 16 months, respectively. Of the 4 HPV-negative patients with tumors greater than 4 cm, 3 are disease-free at 41, 42, and 3 months, respectively, and 1 was lost to follow-up. The third HPV-positive patient had a tumor less than 1 cm, and is disease-free after 38 months. While the number of HPV-positive cases is too small for definitive conclusions, it is possible that for large tumors the presence of HPV DNA may portend a worse prognosis.(ABSTRACT TRUNCATED AT 250 WORDS)

Tumor-associated glycoprotein distribution detected by monoclonal antibody B72.3 in salivary neoplasia.

The expression of tumor-associated glycoprotein (TAG-72), an oncofetal mucin-like tumor-associated glycoprotein derived from membrane-enriched fractions of metastatic breast carcinoma, has been detected by monoclonal antibody (MoAb) B72.3 in adenocarcinomas of breast, colon, lung, endometrium, pancreas, and ovary. The authors reported the scope of TAG-72 expression detected by MoAb B72.3 in salivary neoplasia. They examined 96 salivary lesions (53 malignant and 37 benign primary tumors, 2 metastatic carcinomas, and 4 other benign lesions) and 17 normal tissues from parotid glands and found: diffuse TAG-72 expression in 29 of 55 (53%) malignant tumors and 6 of 36 (17%) benign tumors and in no normal tissue; focal TAG-72 expression in 10 of 55 (17%) malignant salivary tumors, 10 of 37 (25%) benign salivary tumors (all benign mixed tumors), and 1 of 17 (6%) histologically normal parotid gland ducts. Any expression of TAG-72, whether diffuse or focal, was found to have a 71% sensitivity for detecting salivary malignant tumors, but an unacceptably low specificity for malignant lesions (57%). Alternatively, if only diffuse TAG-72 expression was regarded as indicative of malignancy, the specificity of diffuse TAG-72 expression was 86%, but sensitivity of detection decreased to 53%. The authors studied a subset of benign and malignant mixed tumors (BMT and MMT) and found that 12 of 15 (80%) MMT diffusely and strongly expressed TAG-72, 2 of 15 MMT (13%) expressed TAG-72 focally, and 1 MMT (7%) was nonreactive. By contrast, most BMT did not express TAG-72; only sparse, focal TAG-72 expression was seen in 10 of 27 (37%) BMT. If diffuse TAG-72 expression is considered indicative of malignancy, its sensitivity and specificity for malignant mixed tumors is 80% and 100%, respectively. The authors suggest that diffuse TAG-72 expression may resolve conflicts in determining whether or not a mixed tumor is malignant.

Inflammatory pseudotumor of the parapharyngeal space: case report and review of the literature.

Inflammatory pseudotumor of the upper airway is an uncommon lesion of unknown etiology, clinically mimicking a neoplastic process. We document a case of inflammatory pseudotumor of the parapharyngeal space, occurring in a patient with history of cocaine abuse. Corticosteroid treatment was successful in reducing the symptoms. The difficulties in establishing this clinicopathologic diagnosis are discussed and the pertinent literature is reviewed.

Oncocytic tumors of major salivary glands. A study of 68 cases with follow-up of 44 patients.

Oncocytic tumors rarely occur in major salivary glands and generally account for less than 1% of all salivary tumors. Until now, a large series of these tumors with long-term follow-up has been lacking. We report on 68 cases of oncocytic major salivary gland tumors with clinical data on 44 patients. Eighty-four percent occurred in the parotid (male to female ratio of 1:1), and 11% arose in the submandibular gland (six males, one female). Additionally, 5% were incidentally found in salivary nests of the upper cervical lymph nodes. The mean age of all patients was 58 years. Unexpectedly, 20% of the patients had either radiotherapy to the face or upper torso or long-term occupational radiation exposure, 5-40 years prior to tumor discovery. Patients with previous radiation exposure had a mean age of 43 years at tumor discovery as compared with 63 years for all other patients (p less than 0.01). Among 44 patients, there were four definite, documented cases of recurrences: two were multiple and bilateral. A minimum 7% incidence of bilateral disease was noted. An association exists among bilateral disease, tumor recurrence, and extensive clear cell change ("clear cell oncocytosis"). Only one case metastasized: an oncocytic adenocarcinoma of the submandibular gland. None of the oncocytomas studied (including three with perineural spread) metastasized after 0.5-38 years' follow-up (mean 12 years). The literature is reviewed with regard to cases of metastasizing malignant oncocytomas.

Salivary duct carcinoma (cribriform salivary carcinoma of excretory ducts). A clinicopathologic and immunohistochemical study of 12 cases.

Salivary duct carcinoma (cribriform salivary carcinoma of the excretory ducts [CSCED]) is an uncommon malignant tumor which occurs predominantly in men (83% in this series; mean age, 61 years) and most often in the parotid gland (92% in this series). The outcome is unfavorable for most patients; of 11 of 12 patients with follow-up, 45% had local recurrence, 54% had distant metastasis, and 45% were dead of disease within 10 years of diagnosis (mean, 3 years). Metastases to lymph nodes were common (72%). Immunohistochemical studies on paraffin-embedded tissue revealed that most tumors reacted with antibodies known to mark adenocarcinoma: B72.3 (11 of 11) and Lewis Y (ten of ten). High and low molecular weight cytokeratins were present in most tumors (nine of ten and seven of nine cases, respectively), supporting the concept that these adenocarcinomas were of ductal origin. Parotid ducts adjacent to CSCED expressed B72.3 in six of nine cases studied, but parotid ducts from normal tissue (adjacent to benign mixed tumors or enlarged periparotid lymph nodes) rarely expressed this marker (one of 17 cases). The detection of B72.3 diffusely in parotid ducts, especially those with atypia, may imply the presence of malignant tumor nearby, which could be useful in evaluating limited tissue from the parotid. However, further studies are necessary to confirm the significance of this finding.

Supraglottic hemangioma during pregnancy.

A 37-year-old woman, gravida 10, para 1-0-8-1, presented in her second trimester of pregnancy with complaints of frequent sore throats, change in voice quality, and recumbent dyspnea. These symptoms were noted initially during her previous pregnancy, and resolved after delivery. She was found to have a supraglottic hemangioma which occluded 40% of her upper airway. This lesion was noted to dramatically involute beginning at 36 weeks' gestation. The supraglottic hemangioma is a very rare cause of dyspnea in pregnancy. Its growth during pregnancy and regression postpartum is possibly related to the effects of estrogen and progesterone.